One Nation, many preferences

Observations on the form of One Nation preferences, as pollsters and aggregators grapple with the question of how to distribute them.

NOTE: Can the comments thread for this post be reserved for discussion of the matter at hand. The “main” discussion thread is here.

UPDATE: I have a piece on Newspoll and One Nation preferences in Crikey today. Unfortunately, my assertion that the only public statement on the matter had consisted of two tweets from Galaxy on Thursday was superseded by events by the time it was published – namely, that a report by Ben Packham of The Oz today reveals that Newspoll did, indeed, change its allocation of One Nation preferences in December (and not just before the most recent result, as per Barrie Cassidy of the ABC last week). However, the exact split of One Nation preferences being used by Newspoll remains a trade secret.

The subject of One Nation preferences has dominated opinion poll discussion ever since Newspoll came out with its surprise 51-49 result last week, so I thought I’d offer my own contribution by observing their behaviour at recent elections, known or surmised. To start with the former, we have hard data on how One Nation preferences divided between the Coalition and Labor at the five federal elections going back to 2004, and at the Western Australian state election last year:

To L-NP To ALP  
WA 2017 17252 60.6% 11196 39.4% 29 out of 57
Federal 2016 88327 50.5% 86693 49.5% 15 out of 150
Federal 2013 12147 55.1% 9899 44.9% 15 out of 150
Federal 2010 14894 54.8% 12290 45.2% 21 out of 150
Federal 2007 17303 53.0% 15347 47.0% 35 out of 150
Federal 2004 78941 56.4% 61015 43.6% 77 out of 150

The party was a weak presence between 2004 and 2013: it scored only 1.2% of the national vote when contesting over half the seats in 2004, then contested only a scattering of seats in 2007, 2010 and 2013. Nonetheless, its voters were highly consistent throughout this period in leaning modestly in favour of the Coalition. Then came the great One Nation revival in 2016 election, at which, sadly for pollsters and prognosticators, it only offered 15 candidates as a test of its preference behaviour. A seat by seat account goes as follows, with notes added for the two seats where they directed preferences to a particular party rather than running a split ticket, and the two where they were top of the ballot paper:

To L-NP   To ALP    
Blair (Qld) 6603 49.7% 6670 50.3% HTV to L-NP
Dobell (NSW) 3677 44.2% 4649 55.8%  
Fadden (Qld) 5684 53.2% 5009 46.8%  
Fairfax (Qld) 4887 54.3% 4119 45.7%  
Flynn (Qld) 7467 50.0% 7481 50.0%  
Herbert (Qld) 5777 48.3% 6173 51.7%  
Hinkler (Qld) 8679 51.1% 8308 48.9%  
Leichhardt (Qld) 3705 54.7% 3070 45.3%  
Longman (Qld) 3608 43.5% 4685 56.5% HTV to ALP
Maranoa (Qld) 9296 57.9% 6751 42.1%  
Oxley (Qld) 3643 51.9% 3380 48.1% Donkey vote to LNP
Paterson (NSW) 4735 36.3% 8321 63.7% Donkey vote to ALP
Richmond (NSW) 3384 54.9% 2776 45.1%  
Wide Bay (Qld) 7265 51.8% 6757 48.2%  
Wright (Qld) 9917 53.7% 8544 46.3%

For whatever reason, the traditional Coalition lean disappeared on this occasion, with preferences splitting straight down the middle. Kevin Bonham points to the unusual circumstance of the party’s determination to oust Wyatt Roy in Longman, but excluding this result only increases the Coalition’s preference share from 50.5% to 50.8%. The bigger outlier is the extent of the flow to Labor in Paterson, which is presumably the effect of the donkey vote.

The reason a pollster might hesitate to accept this even split will be a permanent state of affairs is the failure of the last two state elections contested by One Nation to bear it out. As the top table shows, One Nation preferences clearly broke to the Liberals in Western Australia, no doubt reflecting the preference deal between the two, politically disastrous though it may have been. However, the result was even more pronounced in Queensland, albeit that we do not have hard data on preference flows to go on here.

What we can do though is try to model the effect of the One Nation vote on preference flows using the results of the thirty-eight seats that had One Nation candidates and Labor-versus-LNP final two-party counts. This is a particularly useful election for this exercise, as One Nation directed preferences in all seats on this occasion, and divided their recommendations fairly evenly between Labor and the LNP. It thus provides an incidental opportunity to measure the impact of their how-to-vote cards.

Against a dependent variable that records Labor’s share of preferences in each electorate, the model regresses a variable measuring One Nation’s share of the minor party vote (ONPshare); a variable with a value of 1 in seats where the One Nation how-to-vote card had Labor ahead of the LNP, and -1 for vice-versa (ONPdummy); and a variable that records Labor’s share of the major party vote, so as to record the natural tendency of preferences to favour the major party that is stronger in a given area (ALPvote):

Since all results are highly significant, we can surmise that the LNP would typically get 65.6% of One Nation preferences where the two major parties recorded the same primary vote and how-to-vote cards weren’t a factor (the intercept plus ONPshare plus half the value of ALPvote). Beyond that, the how-to-vote card could be expected to swing the total 4.6% in one direction or the other.

Given twelve of the fifteen seats contested by One Nation at the federal election were in Queensland, the very substantial difference between the two results is at least a little perplexing. However, by far the likeliest explanation is that One Nation’s gains since 2016 have been mostly Coalition supporters who are likely to feed their preferences back to their own party, if they do indeed end up voting for One Nation.

Adding further to the uncertainty is the tendency shown by One Nation, and minor parties more generally, to shed support during election campaigns – partly because of the shambolic spectacle they usually present during the campaign, and partly because they tend not to contest all the seats. To the extent that One Nation is punching above its weight in recording 7.5% in the BludgerTrack poll aggregate, there will be a corresponding understatement of real-world support for other parties – perhaps particularly in the case of the Coalition, who may end up recovering some of that “soft” One Nation support.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

32 comments on “One Nation, many preferences”

  1. It seems to me there was a bit of a protest vote going to ON from conservative LNP voters, exacerbated by them losing their champion (Abbott).

    At the WA election ON was on about 10% (nationally), since then they have lost about a quarter of that, so perhaps those extra 2.5% of voters who stuck with them in 2017 have gone back to LNP…

    So I wonder if the ON primary vote should be taken into consideration when determining preference flows, upto a certain amount traditional preference flows should be used, but beyond a certain threshold a higher preference to LNP should be used.

    As to actual numbers, i have no idea….

  2. I believe that government incumbency, regardless of whether One Nation choose to direct preferences against incumbents again, will have a big impact. A lot of these voters are quite possibly just going to vote against the government, be it a State Labor government or a Federal LNP Government.

    That’s definitely one way that I reconcile a near 50-50 split of preferences in Hinkler (even with a split ticket) with preferences in its state seats in 2017 which went ~70-30 (Hervey Bay and Bundaberg, HTV preferenced LNP) and 60-40(Burnett, ALP preferenced on HTV).

    So my expectation would be that the “natural preference rate” 65.6% to the LNP number derived from the state election will be a fair bit lower in the Federal election.

  3. One other thing to look at (again I used wide bay because it is a bit weird) is the tendency for minor parties parties to exchange preferences.

    So Katter was tiny. 43% went to LNP, oddly 8% to the Greens, 22% to Labor. This is possibly predicable – small number of voters etc.

    Family first is more bizarre with only 24% going to the LNP and 21% going to the greens. Weird. OK still small numbers

    The Glen Lazarus team seemed more unusual with 54% going ON, more to the Greens and only 13% to the LNP (20% or so of his preferences had come in form the other two minors FF and Katter)

    However when we get to the Greens vote 19% each went to PHON and to LNP. Since 15% came from other minors including Right wing minors such as FF I guess this is to be expected, but still quite surprising.

    Then the final ON split it went 52% to ALP (seems to be a difference with William’s figures that are no doubt more accurate but I can only use the stuff on the AEC website). Now of course about 10% of the ON vote cam via the Greens so it is to be expected to go ALP and also a goodly share of the Lazarus votes (about 15% of the ON)

    Anyway as I sid just a bit unpredictable

    A former long standing MP I know said to me that the vote for BOTH majors in Qld is very soft. That gels with my gut feel. I am not sure about the rest of the country, but have a gut feel that in Victoria it is more solid.

    Queensland is of course always a bit weird.

    I think for example colourful or popular MP (Christiansen and Entch) will buck any national swing against the LNP.

  4. The Government’s dog-whistling (Captain Cook memorial, South African immigration etc) will also drive some One Nation preferences towards the Coalition at the next election.

    There might be more of a feeling by ONP voters that they’re voting ‘for’ Dutton or Morrison and not ‘for’ Turnbull.

  5. Based on the above, at a national level, Newspoll’s apparent assumption that 60% of ON preferences go to the Coalition looks a bit high. 60% might be reasonable in Qld, but looks like it should be 50/50 elsewhere. Am I missing something?

  6. So, Federally it looks like the PHON sends preferences roughly 50-50% to both major parties. This is good news for Labor as they enjoy a much higher second preferences flow from the Greens.
    The Qld 2016 Federal anomaly seems to be over… This Coalition Federal Government is gone!

  7. YouGov apparently does not understand the first law of Australian psephology: they’re all mad in Queensland!

  8. Don’t mean to sound like a conspiracy theorist or poll-denier but how Newspoll has conducted this whole thing is very dodgy. They should have been more transparent.

    It appears at low levels ON vote is evenly distributed. When ON starts to increase primary vote preferences will skew to Coalition as most support is drawn from Coalition voters. Wouldn’t that mean preference flow from One Nation to Coalition will reduce as One Nation voters return back to Coalition?

    Also, Newspoll should come clean on how state election ON preferences get distributed in VIC and NSW. There was a 51-49 poll in Victoria which looked more like 52-48 and now I wonder if ON is flowing the same as federal calculations.

  9. I know that, as KB has pointed out, these changes came in a week or two before YouGov took over Newspoll. But they were surely introduced with the knowledge and consent of the prospective buyer. The extend to which YouGov drove them is, of course, very difficult to say.

  10. Do we know whether the old hands at Newspoll are still there (the stalwarts) or were cleaned out when YouGov moved in. Or did they take the money and run?

  11. For One Nation in Queensland 2017 I extracted data from the preference distributions for those seats where PHON were eliminated before both ALP and L-NP. The split was LNP 145169 – ALP 77619 (65.1% so virtually identical to William’s model). That includes preferences received from other candidates and passed on by One Nation (a significant factor in, eg, the flow to ALP in Redlands), and excludes votes that flowed to non-major party candidates where ON were eliminated fourth or worse. It also excludes the 183412 votes held by One Nation at a point where either Labor or LNP were eliminated from a count.

  12. William

    My stats are very rusty (actually never learned except in passing) so can you explain how you got to the 65% figure – sorry for being dense.

  13. @Kevin Bonham

    The preference distribution you have calculated for One Nation during the last Queensland state election will be replicated at the forthcoming federal election.

    I get a feeling conservative voters defecting from the Coalition to One Nation, Australian Conservatives and the Australian Liberty Alliance would prefer a Coalition over a Labor government, just without Malcolm Turnbull leading it. If the Coalition get re-elected at the forthcoming election Malcolm Turnbull will face a leadership challenge sometime in the following term, with either Scott Morrison or Peter Dutton potentially replacing Malcolm Turnbull.

  14. I would expect there would be something of a “turnbull effect” amongst many ON voters – causing more preferences to skew to labor, than if someone more conservative like Abbott was leading the libs.

  15. Adrian – Yep, a lot of ON voters would not be fans of “Mr Harbourside Mansion”. They know a toff when they see one. That may be a simple reason why the Qld state polling won’t be replicated in a federal election.

  16. DTT, there are four correlation coefficients listed under the “estimate” column. One Nation’s typical preference flow in a given seat will equal the sum of a) the intercept (0.594); b) the ONPshare coefficient (-0.358) multiplied by One Nation’s share of the minor party vote; c) the ONPdummy coefficient (0.046) multipled by 1 if they direct preferences to Labor, -1 if they direct preferences against them, and zero if they do neither; and d) the ALPshare coefficient (0.217) multipled by Labor’s share of the major party vote. In a hypothetical case where One Nation are the only minor party in the field, the number we will multiply ONPshare by will be one. Where the two major parties have the same primary vote, we will multiply ALP share by 0.5. So:

    0.594 + (-0.358 * 1) + (0.046 * 0) + (0.217 * 0.5)
    0.594 – 0.358 + 0 + 0.1085

    So Labor gets 34.45% of preferences and the LNP gets 65.55%.

  17. Sometimes in the media people say for example “80% of Greens preferences go to Labor so Labor get an 8% boost from a 10% Greens vote”, which I suppose is strictly true in one sense. I prefer to look at it in that example that Labor get 80% of 10% = 8% for them and 20% of 10% = 2% against them, so really a “boost” to their TPP of 6%.

    As the Greens vote and preference distribution don’t change that much, the One Nation version is much more interesting. Because say ON get 10% and 60% go to the Coalition, the TPP effect is only 6-4 = 2%. This is crucial because one would imagine if there were truly only two alternatives on the ballot paper (say a run-off vote for President like in France) a higher proportion of those ON voters would vote Coalition, say 75-80% which would boost the Coalition TPP vote by 5-6% instead.

    I believe that ON’s presence has helped Labor win 4 elections – both Queensland & WA in 2001 and 2017. Can they add a Federal Election to their tally?

  18. @Rocket Rocket

    You raise some good points here, I would argue that One Nation preferences will generally spilt roughly 60-40% the Coalition’s way. However that would vary by seat if One Nation decides to preference against both Labor and Coalition sitting members. Because One Nation voters are for the greater part former Coalition voters, however a sizeable proportion are former Labor voters. In hindsight I would argue One Nation is the closest we have in Australia to parties such as the Party for Freedom in the Netherlands, Alternative for Germany and the Swedish Democrats among others. It is notable that One Nation does not advocate Economic neoliberalism like the parties I will mention below.

    With the Australian Conservatives, Rise Up Australia, Liberal Democratic Party and the Australian Liberty Alliance. I would argue their voters come from the Conservative wing of the Liberal Party and will preference the Coalition strongly like the Greens do to Labor. All together could get something like between 3-4% of the vote nationally come the election. Unlike One Nation voters the voters of those parties prefer a Coalition government although without Turnbull in charge, a Labor government is an anathema to them.

  19. William, I can think of at least three reasons why the Libs will not get anywhere near 65 percent of ON preferences at the next election:
    (a) the Libs have a merchant banker toff in charge. He is rat-poison to many ON voters;
    (b) the Libs are the incumbent government. ON voters are sore with the world. They are protest voters. They hate incumbents;
    ( c) Queensland ain’t Australia. As Jake Gittes said about Chinatown: they do things different there.

  20. Prior to the 2016 federal election, have state election preference allocations historically proven to be a reliable indicator of subsequent federal preference allocations? I’d doubt it. If state elections are showing a strong LNP-leaning then it sounds like good cause to assume the historic federal high water mark, not take the state levels verbatim.

  21. One thing that many may not realise about many of the Qld seats is that they are heavily divided with large contingents of BOTH Green and Phon. This I suspect is unusual. Basically it is a phenomenon of the mainly northern rural hinterland and the Sunshine Coast. The most citified is Dickson which oddly enough has the booth with the highest Green vote in Qld (possibly Australia) which at 49% is high. It is of course a small (but not tiny) booth. At the same time while Phon did not run last election there are pockets where I would expect a very large One Nation vote -Kurwongbah for example.

    There are quite a lot of spots like this all over those seats – Fairfax, Wide Bay come to mind first.

  22. William, I know I spout a lot of crap (let’s not harp on it) but according to wikipedia, one nation got 13.73 per cent of the primary vote in Queensland at the last election! Yet, federally it is polling about half of that (7.5 per cent). Surely, those are the two most important figures to keep in mind.
    The most logical explanation for the 65 per cent preference allocation to the libs in the Qld election is that (as Bug1 and other state above) many natural libs voted one nation first and then sent their votes home.
    But bedrock One Nation supporters (the 7.5 per cent in the federal polling) have a very different preference profile (as previous federal elections have shown).
    I really don’t understand how Newspoll can use only one data point (a Qld election where there was a very different dynamic to a federal poll (that is obvious from the fact that One Nation polled 13.73) to change its methodology. Looks pretty heroic to me. And I don’t mean that as a compliment.

  23. How can the 7.5% in federal polling be “bedrock One Nation supporters”? The party has only ever polled that well in 1998, and certainly did not do so in 2016. I cannot imagine why you would conclude that the voters gained by One Nation at the Queensland election were “natural Libs”, but the federal ones aren’t.

  24. One Nation is a right winged party which gives most of its support to other right winged parties in Australia. Given that, there are three well known right winged parties in Australia.

  25. One Nation appeals to those with a racial bent of a low sociological order who are predisposed to blaming those they deem responsible for their own demise – nut jobs who lack the ability to read, write and reason – I suggest they pick up a book and read it, write a nice letter and think about what they can best do to help society as a whole – Itt will solve so many of their own woes

Leave a Reply

Your email address will not be published. Required fields are marked *